A power system is configured by many power generators, loads, power transmission and distribution apparatuses, and control devices. Slight disturbance due to load variation, system switching, and the like is added to the power system in a normal state. When an accident such as ground fault or a short circuit of a system occurs, large disturbance is added to the power system. Power fluctuation with a plurality of fluctuation components is caused by the disturbance.
In a case where fluctuation mode components are small or are attenuated at a high speed, the system fluctuation is retained in a sufficiently small range after the disturbance occurs, and thus, the power system is in a stable state. However, in a case where the fluctuation is large or attenuation is continued or distributed without convergence, a power generator stops or wide power failure occurs, as a result. Accordingly, there is a possibility that stable management cannot be performed.
There is a tendency that as a power system scale increases, power source capacity increases, and remoting, localization, lengthening of a power transmission wire according to the localization, and a heavy flow are performed. Since effects according to the change occur in a direction in which stability of a power system decreases, there is a possibility that knock-on effects to a large-scale power-off and wide power failure being started from step-out of a power generator at the time of system failure increases.
Accordingly, an on-line system stability evaluating method of normally monitoring system stability and of preventing and predicting spreading to wide power failure in advance is required.
Recently, a phase detector PMU (Phasor Measurement Unit) which measures a bus voltage phase angle of a power system in an approximately real time by using synchronous measurement which uses a GPS (Global Positioning System) has been used. It is possible to grasp a power fluctuation phenomenon in real time by using an on-line measurement value such as a system bus voltage measured by the phase detector PMU.
A system stability monitoring technology that an operator which manages a power system uses to monitor a state of the system has been developed on the basis of the aforementioned background. For example, PTL 1 describes “is characterized by including a phase measuring device that measures an amount and a phase angle of a system bus voltage of each subsystem provided in each of a plurality of subsystems categorized such that analysis of a power fluctuation mode of a power system can be performed, data input processing means for inputting an amount and a phase angle of a system bus voltage for each subsystem which is measured by the phase measuring device in each subsystem other than a reference in time series by using any one of the plurality of subsystems as the reference and for obtaining phase angle deviation and frequency deviation which use a phase angle and a frequency of a reference subsystem as a reference on the basis of the phase angle, an observation data storage unit that stores phase angle deviation and frequency deviation of a system bus voltage for each subsystem obtained by the data input processing means as observation data, observation data take-out means for randomly taking out observation data of each subsystem from a plurality of subwindows which have widths smaller than a width of a main window with a predetermined same period width with respect to observation data of the observation data storage unit and have size different from each other, state transition equation writing means for writing a state transition equation which uses phase angle deviation and frequency deviation of observation data taken out by the observation data take-out means as state variables in each subwindow, characteristic value calculation means for obtaining a characteristic value of a coefficient matrix of the state transition equation in each subwindow written by the state transition equation writing means, and output processing means for displaying and outputting a distribution of a characteristic value group in each subwindow calculated by the characteristic value calculation means as a complex plane”.